Lymphatic Uptake of a Highly Lipophilic Protease Inhibitor Prodrug from a Lipid-Based Formulation is Limited by Instability in the Intestine.

Dabigatran etexilate (DABE) is a lipophilic double alkyl ester prodrug of dabigatran (DAB) which is a serine protease inhibitor used clinically as an anticoagulant. Recently, translocation of serine protease enzymes, including trypsin, from the gut into the mesenteric lymph and then blood has been associated with organ failure in acute and critical illnesses (ACIs). Delivery of DABE into mesenteric lymph may thus be an effective strategy to prevent organ failure in ACIs. Most drugs access the mesenteric lymph in low quantities following oral administration, as they are rapidly transported away from the intestine via the blood. Here, we examine the potential to deliver DABE into the mesenteric lymph by promoting association with lymph lipid transport pathways via co-administration with a lipid-based formulation (LBF). A series of self-emulsifying LBFs were designed and tested in vitro for their potential to form stable DABE loaded emulsions and keep DABE solubilised and stable over time in simulated gastrointestinal conditions. The LBFs were found to form fine emulsions with a droplet size of 214 ± 30 nm and DABE was stable in the formulation. The stability of DABE in vitro in simulated intestinal conditions, plasma and lymph samples was also evaluated to ensure stability in collected samples and to evaluate whether the prodrug is likely to release active DAB. Ultimately, a highly uniform and stable self-emulsifying Type III A LBF of DABE was chosen for progression into in vivo studies in male Sprague Dawley rats to confirm the lymphatic uptake and plasma pharmacokinetics. Both in vitro and in vivo in plasma and lymph, DABE was rapidly converted to an intermediate and DAB. The main species present in vivo in both plasma and lymph was DAB and mass transport of DABE and DAB in lymph was minimal (∼0.5 % of dose). Importantly, the concentration of DABE in lymph was substantially (20-176 fold) higher than in plasma, supporting that if the prodrug were stable and did not convert to DAB in the intestine, it would be lymphatically transported. Future studies will therefore focus on optimizing the design of the prodrug and formulation to improve stability during absorption and further promote lymphatic uptake.

Extracting social determinants of health events with transformer-based multitask, multilabel named entity recognition.

OBJECTIVE: Social determinants of health (SDOH) are nonclinical, socioeconomic conditions that influence patient health and quality of life. Identifying SDOH may help clinicians target interventions. However, SDOH are more frequently available in narrative notes compared to structured electronic health records. The 2022 n2c2 Track 2 competition released clinical notes annotated for SDOH to promote development of NLP systems for extracting SDOH. We developed a system addressing 3 limitations in state-of-the-art SDOH extraction: the inability to identify multiple SDOH events of the same type per sentence, overlapping SDOH attributes within text spans, and SDOH spanning multiple sentences. MATERIALS AND METHODS: We developed and evaluated a 2-stage architecture. In stage 1, we trained a BioClinical-BERT-based named entity recognition system to extract SDOH event triggers, that is, text spans indicating substance use, employment, or living status. In stage 2, we trained a multitask, multilabel NER to extract arguments (eg, alcohol "type") for events extracted in stage 1. Evaluation was performed across 3 subtasks differing by provenance of training and validation data using precision, recall, and F1 scores. RESULTS: When trained and validated on data from the same site, we achieved 0.87 precision, 0.89 recall, and 0.88 F1. Across all subtasks, we ranked between second and fourth place in the competition and always within 0.02 F1 from first. CONCLUSIONS: Our 2-stage, deep-learning-based NLP system effectively extracted SDOH events from clinical notes. This was achieved with a novel classification framework that leveraged simpler architectures compared to state-of-the-art systems. Improved SDOH extraction may help clinicians improve health outcomes.

Increasing Linker Chain Length and Intestinal Stability Enhances Lymphatic Transport and Lymph Node Exposure of Triglyceride Mimetic Prodrugs of a Model Immunomodulator Mycophenolic Acid.

Targeted delivery of immunomodulators to the lymphatic system has the potential to enhance therapeutic efficacy by increasing colocalization of drugs with immune targets such as lymphocytes. A triglyceride (TG)-mimetic prodrug strategy has been recently shown to enhance the lymphatic delivery of a model immunomodulator, mycophenolic acid (MPA), via incorporation into the intestinal TG deacylation-reacylation and lymph lipoprotein transport pathways. In the current study, a series of structurally related TG prodrugs of MPA were examined to optimize structure-lymphatic transport relationships for lymph-directing lipid-mimetic prodrugs. MPA was conjugated to the sn-2 position of the glyceride backbone of the prodrugs using linkers of different chain length (5-21 carbons) and the effect of methyl substitutions at the alpha and/or beta carbons to the glyceride end of the linker was examined. Lymphatic transport was assessed in mesenteric lymph duct cannulated rats, and drug exposure in lymph nodes was examined following oral administration to mice. Prodrug stability in simulated intestinal digestive fluid was also evaluated. Prodrugs with straight chain linkers were relatively unstable in simulated intestinal fluid; however, co-administration of lipase inhibitors (JZL184 and orlistat) was able to reduce instability and increase lymphatic transport (2-fold for a prodrug with a 6-carbon spacer, i.e., MPA-C6-TG). Methyl substitutions to the chain resulted in similar trends in improving intestinal stability and lymphatic transport. Medium- to long-chain spacers (C12, C15) between MPA and the glyceride backbone were most effective in promoting lymphatic transport, consistent with increases in lipophilicity. In contrast, short-chain (C6-C10) linkers appeared to be too unstable in the intestine and insufficiently lipophilic to associate with lymph lipid transport pathways, while very long-chain (C18, C21) linkers were also not preferred, likely as a result of increases in molecular weight reducing solubility or permeability. In addition to more effectively promoting drug transport into mesenteric lymph, TG-mimetic prodrugs based on a C12 linker resulted in marked increases (>40 fold) in the exposure of MPA in the mesenteric lymph nodes in mice when compared to administration of MPA alone, suggesting that optimizing prodrug design has the potential to provide benefit in targeting and modulating immune cells.

Corrigendum: Triglyceride-mimetic prodrugs of buprenorphine enhance oral bioavailability via promotion of lymphatic transport.

[This corrects the article DOI: 10.3389/fphar.2022.879660.].

Triglyceride-Mimetic Prodrugs of Buprenorphine Enhance Oral Bioavailability via Promotion of Lymphatic Transport.

Buprenorphine (BUP) is a potent opioid analgesic that is widely used for severe pain management and opioid replacement therapy. The oral bioavailability of BUP, however, is significantly limited by first-pass metabolism. Previous studies have shown that triglyceride (TG) mimetic prodrugs of the steroid hormone testosterone circumvent first-pass metabolism by directing drug transport through the intestinal lymphatics, bypassing the liver. The current study expanded this prodrug strategy to BUP. Here different self-immolative (SI) linkers were evaluated to conjugate BUP to the 2 position of the TG backbone via the phenol group on BUP. The SI linkers were designed to promote drug release in plasma. Lipolysis of the prodrug in the intestinal tract was examined via incubation with simulated intestinal fluid (SIF), and potential for parent drug liberation in the systemic circulation was evaluated via incubation in rat plasma. Lymphatic transport and bioavailability studies were subsequently conducted in mesenteric lymph duct or carotid artery-cannulated rats, respectively. TG prodrug derivatives were efficiently transported into the lymphatics (up to 45% of the dose in anaesthetised rats, vs. less than 0.1% for BUP). Incorporation of the SI linkers facilitated BUP release from the prodrugs in the plasma and in concert with high lymphatic transport led to a marked enhancement in oral bioavailability (up to 22-fold) compared to BUP alone. These data suggest the potential to develop an orally bioavailable BUP product which may have advantages with respect to patient preference when compared to current sublingual, transdermal patch or parenteral formulations.

Antibody drug conjugate: the "biological missile" for targeted cancer therapy.

Antibody-drug conjugate (ADC) is typically composed of a monoclonal antibody (mAbs) covalently attached to a cytotoxic drug via a chemical linker. It combines both the advantages of highly specific targeting ability and highly potent killing effect to achieve accurate and efficient elimination of cancer cells, which has become one of the hotspots for the research and development of anticancer drugs. Since the first ADC, Mylotarg® (gemtuzumab ozogamicin), was approved in 2000 by the US Food and Drug Administration (FDA), there have been 14 ADCs received market approval so far worldwide. Moreover, over 100 ADC candidates have been investigated in clinical stages at present. This kind of new anti-cancer drugs, known as "biological missiles", is leading a new era of targeted cancer therapy. Herein, we conducted a review of the history and general mechanism of action of ADCs, and then briefly discussed the molecular aspects of key components of ADCs and the mechanisms by which these key factors influence the activities of ADCs. Moreover, we also reviewed the approved ADCs and other promising candidates in phase-3 clinical trials and discuss the current challenges and future perspectives for the development of next generations, which provide insights for the research and development of novel cancer therapeutics using ADCs.

Classifying social determinants of health from unstructured electronic health records using deep learning-based natural language processing.

OBJECTIVE: Social determinants of health (SDOH) are non-medical factors that can profoundly impact patient health outcomes. However, SDOH are rarely available in structured electronic health record (EHR) data such as diagnosis codes, and more commonly found in unstructured narrative clinical notes. Hence, identifying social context from unstructured EHR data has become increasingly important. Yet, previous work on using natural language processing to automate extraction of SDOH from text (a) usually focuses on an ad hoc selection of SDOH, and (b) does not use the latest advances in deep learning. Our objective was to advance automatic extraction of SDOH from clinical text by (a) systematically creating a set of SDOH based on standard biomedical and psychiatric ontologies, and (b) training state-of-the-art deep neural networks to extract mentions of these SDOH from clinical notes. DESIGN: A retrospective cohort study. SETTING AND PARTICIPANTS: Data were extracted from the Medical Information Mart for Intensive Care (MIMIC-III) database. The corpus comprised 3,504 social related sentences from 2,670 clinical notes. METHODS: We developed a framework for automated classification of multiple SDOH categories. Our dataset comprised narrative clinical notes under the "Social Work" category in the MIMIC-III Clinical Database. Using standard terminologies, SNOMED-CT and DSM-IV, we systematically curated a set of 13 SDOH categories and created annotation guidelines for these. After manually annotating the 3,504 sentences, we developed and tested three deep neural network (DNN) architectures - convolutional neural network (CNN), long short-term memory (LSTM) network, and the Bidirectional Encoder Representations from Transformers (BERT) - for automated detection of eight SDOH categories. We also compared these DNNs to three baselines models: (1) cTAKES, as well as (2) L2-regularized logistic regression and (3) random forests on bags-of-words. Model evaluation metrics included micro- and macro- F1, and area under the receiver operating characteristic curve (AUC). RESULTS: All three DNN models accurately classified all SDOH categories (minimum micro-F1 = 0.632, minimum macro-AUC = 0.854). Compared to the CNN and LSTM, BERT performed best in most key metrics (micro-F1 = 0.690, macro-AUC = 0.907). The BERT model most effectively identified the "occupational" category (F1 = 0.774, AUC = 0.965) and least effectively identified the "non-SDOH" category (F = 0.491, AUC = 0.788). BERT outperformed cTAKES in distinguishing social vs non-social sentences (BERT F1 = 0.87 vs. cTAKES F1 = 0.06), and outperformed logistic regression (micro-F1 = 0.649, macro-AUC = 0.696) and random forest (micro-F1 = 0.502, macro-AUC = 0.523) trained on bag-of-words. CONCLUSIONS: Our study framework with DNN models demonstrated improved performance for efficiently identifying a systematic range of SDOH categories from clinical notes in the EHR. Improved identification of patient SDOH may further improve healthcare outcomes.

Lipophilic Conjugates of Drugs: A Tool to Improve Drug Pharmacokinetic and Therapeutic Profiles.

Lipophilic conjugates (LCs) of small molecule drugs have been used widely in clinical and pre-clinical studies to achieve a number of pharmacokinetic and therapeutic benefits. For example, lipophilic derivatives of drugs are employed in several long acting injectable products to provide sustained drug exposure for hormone replacement therapy and to treat conditions such as neuropsychiatric diseases. LCs can also be used to modulate drug metabolism, and to enhance drug permeation across membranes, either by increasing lipophilicity to enhance passive diffusion or by increasing protein-mediated active transport. Furthermore, such conjugation strategies have been employed to promote drug association with endogenous macromolecular carriers (e.g. albumin and lipoproteins), and this in turn results in altered drug distribution and pharmacokinetic profiles, where the changes can be 'general' (e.g. prolonged plasma half-life) or 'specific' (e.g. enhanced delivery to specific tissues in parallel with the macromolecular carriers). Another utility of LCs is to enhance the encapsulation of drugs within engineered nanoscale drug delivery systems, in order to best take advantage of the targeting and pharmacokinetic benefits of nanomedicines. The current review provides a summary of the mechanisms by which lipophilic conjugates, including in combination with delivery vehicles, can be used to control drug delivery, distribution and therapeutic profiles. The article is structured into sections which highlight a specific benefit of LCs and then demonstrate this benefit with case studies. The review attempts to provide a toolbox to assist researchers to design and optimise drug candidates, including consideration of drug-formulation compatibility.

Analytical validation of GMEX rapid point-of-care CYP2C19 genotyping system for the CHANCE-2 trial.

BACKGROUND AND PURPOSE: Rapid genotyping is useful for guiding early antiplatelet therapy in patients with high-risk nondisabling ischaemic cerebrovascular events (HR-NICE). Conventional genetic testing methods used in CYP2C19 genotype-guided antiplatelet therapy for patients with HR-NICE did not satisfy the needs of the Clopidogrel in High-Risk Patients with Acute Nondisabling Cerebrovascular Events (CHANCE)-2 trial. Therefore, we developed the rapid-genotyping GMEX (point-of-care) system to meet the needs of the CHANCE-2 trial. METHODS: Healthy individuals and patients with history of cardiovascular diseases (n=408) were enrolled from six centres of the CHANCE-2 trial. We compared the laboratory-based genomic test results with Sanger sequencing test results for accuracy verification. Next, we demonstrated the accuracy, timeliness and clinical operability of the GMEX system compared with laboratory-based technology (YZY Kit) to verify whether the GMEX system satisfies the needs of the CHANCE-2 trial. RESULTS: Genotypes reported by the GMEX system showed 100% agreement with those determined by using the YZY Kit and Sanger sequencing for all three CYP2C19 alleles (*2, *3 and *17) tested. The average result's turnaround times for the GMEX and YZY Kit methods were 85.0 (IQR: 85.0-86.0) and 1630.0 (IQR: 354.0-7594.0) min (p<0.001), respectively. CONCLUSIONS: Our data suggest that the GMEX system is a reliable and feasible point-of-care system for rapid CYP2C19 genotyping for the CHANCE-2 trial or related clinical and research applications.

Targeted delivery of mycophenolic acid to the mesenteric lymph node using a triglyceride mimetic prodrug approach enhances gut-specific immunomodulation in mice.

The mesenteric lymph nodes (MLN) are a key site for the generation of adaptive immune responses to gut-derived antigenic material and immune cells within the MLN contribute to the pathophysiology of a range of conditions including inflammatory and autoimmune diseases, viral infections, graft versus host disease and cancer. Targeting immunomodulating drugs to the MLN may thus be beneficial in a range of conditions. This paper investigates the potential benefit of targeting a model immunosuppressant drug, mycophenolic acid (MPA), to T cells in the MLN, using a triglyceride (TG) mimetic prodrug approach. We confirmed that administration of MPA in the TG prodrug form (MPA-TG), increased lymphatic transport of MPA-related species 83-fold and increased MLN concentrations of MPA >20 fold, when compared to MPA alone, for up to 4 h in mice. At the same time, the plasma exposure of MPA and MPA-TG was similar, limiting the opportunity for systemic side effects. Confocal microscopy and flow cytometry studies with a fluorescent model prodrug (Bodipy-TG) revealed that the prodrug accumulated in the MLN cortex and paracortex at 5 and 10 h following administration and was highly associated with B cells and T cells that are found in these regions of the MLN. Finally, we demonstrated that MPA-TG was significantly more effective than MPA at inhibiting CD4+ and CD8+ T cell proliferation in the MLN of mice in response to an oral ovalbumin antigen challenge. In contrast, MPA-TG was no more effective than MPA at inhibiting T cell proliferation in peripheral LN when mice were challenged via SC administration of ovalbumin. This paper provides the first evidence of an in vivo pharmacodynamic benefit of targeting the MLN using a TG mimetic prodrug approach. The TG mimetic prodrug technology has the potential to benefit the treatment of a range of conditions where aberrant immune responses are initiated in gut-associated lymphoid tissues.

Intestinal delivery in a long-chain fatty acid formulation enables lymphatic transport and systemic exposure of orlistat.

Orlistat is a pancreatic lipase (PL) inhibitor that inhibits dietary lipid absorption and is used to treat obesity. The oral bioavailability of orlistat is considered zero after administration in standard formulations. This is advantageous in the treatment of obesity. However, if orlistat absorption could be improved it has the potential to treat diseases such as acute and critical illnesses where PL transport to the systemic circulation via gut lymph promotes organ failure. Orlistat is highly lipophilic and may associate with intestinal lipid absorption pathways into lymph. Here we investigate the potential to improve orlistat lymph and systemic uptake through intestinal administration in lipid formulations (LFs). The effect of lipid type, lipid dose, orlistat dose, and infusion time on lymph and systemic availability of orlistat was investigated. After administration in all LFs, orlistat concentrations in lymph were greater than in plasma, suggesting direct transport via lymph. Lymph and plasma orlistat derivative concentrations were ~8-fold greater after administration in a long-chain fatty acid (LC-FA) compared to a lipid-free, LC triglyceride (LC-TG) or medium-chain FA (MC-FA) formulation. Overall, administration of orlistat in a LC-FA formulation promotes lymph and systemic uptake which may enable treatment of diseases associated with elevated systemic PL activity.

The Impact of Conjugation Position and Linker Chemistry on the Lymphatic Transport of a Series of Glyceride and Phospholipid Mimetic Prodrugs.

Drug delivery to the lymphatic system is gaining increasing attention, particularly in fields such as immunotherapy where drug access to lymphocytes is central to activity. We have previously described a prodrug strategy that facilitates the lymphatic delivery of a model immunomodulator, mycophenolic acid (MPA) via incorporation into intestinal triglyceride transport pathways. The current study explored a series of structurally related glyceride and phospholipid mimetic prodrugs of MPA in an attempt to enhance lymph targeting and to better elucidate the design criteria for lipid mimetic prodrugs. MPA was conjugated to a glyceride or phospholipid backbone at various positions using different spacers employing ester, ether, carbonate and amide bonds. Patterns of prodrug hydrolysis were evaluated in rat digestive fluid, and lymphatic transport and plasma pharmacokinetics were assessed in lymph duct cannulated rats. Prodrugs with different spacers between MPA and the glyceride backbone resulted in up to 70-fold differences in gastrointestinal stability. MPA conjugation at the 2 position of the glyceride backbone and via an ester bond were most effective in promoting lymphatic transport. Phospholipid prodrug derivatives, or glyceride derivatives with MPA attached at the 1 position or when linked via ether, carbonate or amide bonds were poorly incorporated into lymphatic transport pathways.

Intestinal Lymph Flow, and Lipid and Drug Transport Scale Allometrically From Pre-clinical Species to Humans.

The intestinal lymphatic system transports fluid, immune cells, dietary lipids, and highly lipophilic drugs from the intestine to the systemic circulation. These transport functions are important to health and when dysregulated contribute to pathology. This has generated significant interest in approaches to deliver drugs to the lymphatics. Most of the current understanding of intestinal lymph flow, and lymphatic lipid and drug transport rates, comes from in vitro studies and in vivo animal studies. In contrast, intestinal lymphatic transport studies in human subjects have been limited. Recently, three surgical patients had cannulation of the thoracic lymph duct for collection of lymph before and during a stepwise increase in enteral feed rate. We compared these data to studies where we previously enterally administered controlled quantities of lipid and the lipophilic drug halofantrine to mice, rats and dogs and collected lymph and blood (plasma). The collected lymph was analyzed to compare lymph flow rate, triglyceride (TG) and drug transport rates, and plasma was analyzed for drug concentrations, as a function of enteral lipid dose across species. Lymph flow rate, TG and drug transport increased with lipid administration in all species tested, and scaled allometrically according to the equation A = aM E where A is the lymph transport parameter, M is animal body mass, a is constant and E is the allometric exponent. For lymph flow rate and TG transport, the allometric exponents were 0.84-0.94 and 0.80-0.96, respectively. Accordingly, weight normalized lymph flow and TG mass transport were generally lower in larger compared to smaller species. In comparison, mass transport of drug via lymph increased in a greater than proportional manner with species body mass with an exponent of ∼1.3. The supra-proportional increase in lymphatic drug transport with species body mass appeared to be due to increased partitioning of drug into lymph rather than blood following absorption. Overall, this study proposes that intestinal lymphatic flow, and lymphatic lipid and drug transport in humans is most similar to species with higher body mass such as dogs and underestimated by studies in rodents. Notably, lymph flow and lipid transport in humans can be predicted from animal data via allometric scaling suggesting the potential for similar relationships with drug transport.

Lymphatic Uptake of Liposomes after Intraperitoneal Administration Primarily Occurs via the Diaphragmatic Lymphatics and is Dependent on Liposome Surface Properties.

Drugs are commonly administered via the intraperitoneal (IP) route to treat localized infections and cancers in patients and to test drug efficacy and toxicity in preclinical studies. Despite this, there remain large gaps in our understanding of drug absorption routes (lymph vs blood) and pharmacokinetics following IP administration. This is particularly true when drugs are administered in complex delivery systems such as liposomes which are the main marketed formulation for several drugs that are administered intraperitoneally. This study investigated the impact of liposome surface properties (charge and PEGylation) on absorption into lymph and blood, and lymphatic disposition patterns, following IP administration. To achieve this, stable 3H-dipalmitoyl-phosphatidylcholine (DPPC) and 14C-sucrose-radiolabeled liposomes of 100-150 nm diameter with negative, neutral, or positive surface charge, or a PEGylated surface, were prepared and administered intraperitoneally to rats. Radiolabel concentrations were measured in lymph, blood, and lymph nodes (LNs). Lymph was collected from the thoracic lymph duct at either the abdomen (ABD) or the jugular-subclavian junction (JSJ). The lymphatic recovery of the radiolabels was substantially lower after administration in positively charged compared to the neutral, negative, or PEGylated liposomes. Radiolabel recovery was substantially greater (up to 18-fold) in the thoracic lymph collected at the JSJ when compared to that at the ABD, suggesting that liposomes entered the lymphatics at the diaphragm. Consistent with this, the concentration of the liposome labels was substantially higher (up to seven-fold) in mediastinal than in mesenteric LNs. Overall, this study shows how the peritoneal absorption and lymphatic disposition of drugs administered intraperitoneally can be manipulated through a careful selection of the drug delivery system and may thus be optimized to treat localized conditions such as cancers, infections, inflammatory diseases, and acute and critical illness.

Enhancing the Oral Absorption of Kinase Inhibitors Using Lipophilic Salts and Lipid-Based Formulations.

The absolute bioavailability of many small molecule kinase inhibitors (smKIs) is low. The reasons for low bioavailability are multifaceted and include constraints due to first pass metabolism and poor absorption. For smKIs where absorption limits oral bioavailability, low aqueous solubility and high lipophilicity, often in combination with high-dose requirements have been implicated in low and variable absorption, food-effects, and absorption-related drug-drug interactions. The current study has evaluated whether preparation of smKIs as lipophilic salts/ionic liquids in combination with coadministration with lipid-based formulations is able to enhance absorption for examples of this compound class. Lipophilic (docusate) salt forms of erlotinib, gefitinib, ceritinib, and cabozantinib (as example smKIs demonstrating low aqueous solubility and high lipophilicity) were prepared and isolated as workable powder solids. In each case, the lipophilic salt exhibited high and significantly enhanced solubility in lipidic excipients (>100 mg/g) when compared to the free base or commercial salt form. Isolation as the lipophilic salt facilitated smKI loading in model lipid-based formulations at high concentration, increased in vitro solubilization at gastric and intestinal pH and in some cases increased oral absorption (∼2-fold for cabozantinib formulations in rats). Application of a lipophilic salt approach can therefore facilitate the use of lipid-based formulations for examples of the smKI compound class where low solubility limits absorption and is a risk factor for increased variability due to food-effects.

Data and systems for medication-related text classification and concept normalization from Twitter: insights from the Social Media Mining for Health (SMM4H)-2017 shared task.

Objective: We executed the Social Media Mining for Health (SMM4H) 2017 shared tasks to enable the community-driven development and large-scale evaluation of automatic text processing methods for the classification and normalization of health-related text from social media. An additional objective was to publicly release manually annotated data. Materials and Methods: We organized 3 independent subtasks: automatic classification of self-reports of 1) adverse drug reactions (ADRs) and 2) medication consumption, from medication-mentioning tweets, and 3) normalization of ADR expressions. Training data consisted of 15 717 annotated tweets for (1), 10 260 for (2), and 6650 ADR phrases and identifiers for (3); and exhibited typical properties of social-media-based health-related texts. Systems were evaluated using 9961, 7513, and 2500 instances for the 3 subtasks, respectively. We evaluated performances of classes of methods and ensembles of system combinations following the shared tasks. Results: Among 55 system runs, the best system scores for the 3 subtasks were 0.435 (ADR class F1-score) for subtask-1, 0.693 (micro-averaged F1-score over two classes) for subtask-2, and 88.5% (accuracy) for subtask-3. Ensembles of system combinations obtained best scores of 0.476, 0.702, and 88.7%, outperforming individual systems. Discussion: Among individual systems, support vector machines and convolutional neural networks showed high performance. Performance gains achieved by ensembles of system combinations suggest that such strategies may be suitable for operational systems relying on difficult text classification tasks (eg, subtask-1). Conclusions: Data imbalance and lack of context remain challenges for natural language processing of social media text. Annotated data from the shared task have been made available as reference standards for future studies (http://dx.doi.org/10.17632/rxwfb3tysd.1).

Transformation of Biopharmaceutical Classification System Class I and III Drugs Into Ionic Liquids and Lipophilic Salts for Enhanced Developability Using Lipid Formulations.

Higher lipid solubility of lipophilic salt forms creates new product development opportunities for high-dose liquid-filled capsules. The purpose of this study is to determine if lipophilic salts of Biopharmaceutical Classification System (BCS) Class I amlodipine and BCS Class III fexofenadine, ranitidine, and metformin were better lipid formulation candidates than existing commercial salts. Lipophilic salts were prepared from lipophilic anions and commercial HCl or besylate salt forms, as verified by 1H-NMR. Thermal properties were assessed by differential scanning calorimetry and hot-stage microscopy. X-ray diffraction and polarized light microscopy were used to confirm the salt's physical form. All lipophilic salt forms were substantially more lipid-soluble (typically >10-fold) when compared to commercial salts. For example, amlodipine concentrations in lipidic excipients were limited to <5-10 mg/g when using the besylate salt but could be increased to >100 mg/g when using the docusate salt. Higher lipid solubility of the lipophilic salts of each drug translated to higher drug loadings in lipid formulations. In vitro tests showed that lipophilic salts solubilized in a lipid formulation resulted in dispersion behavior that was at least as rapid as the dissolution rates of conventional salts. This study confirmed the applicability of forming lipophilic salts of BCS I and III drugs to promote the utility of lipid-based delivery systems.

Computational Models of the Gastrointestinal Environment. 2. Phase Behavior and Drug Solubilization Capacity of a Type I Lipid-Based Drug Formulation after Digestion.

Lipid-based drug formulations can greatly enhance the bioavailability of poorly water-soluble drugs. Following the oral administration of formulations containing tri- or diglycerides, the digestive processes occurring within the gastrointestinal (GI) tract hydrolyze the glycerides to mixtures of free fatty acids and monoglycerides that are, in turn, solubilized by bile. The behavior of drugs within the resulting colloidal mixtures is currently not well characterized. This work presents matched in vitro experimental and molecular dynamics (MD) theoretical models of the GI microenvironment containing a digested triglyceride-based (Type I) drug formulation. Both the experimental and theoretical models consist of molecular species representing bile (glycodeoxycholic acid), digested triglyceride (1:2 glyceryl-1-monooleate and oleic acid), and water. We have characterized the phase behavior of the physical system using nephelometry, dynamic light scattering, and polarizing light microscopy and compared these measurements to phase behavior observed in multiple MD simulations. Using this model microenvironment, we have investigated the dissolution of the poorly water-soluble drug danazol experimentally using LC-MS and theoretically by MD simulation. The results show how the formulation lipids alter the environment of the GI tract and improve the solubility of danazol. The MD simulations successfully reproduce the experimental results showing the utility of MD in modeling the fate of drugs after digestion of lipid-based formulations within the intestinal lumen.